locked
OWA through UAG problems RRS feed

  • Question

  • Hi there,

    I'm having problems with OWA on Exchange 2010 published behind UAG and TMG.  We seem to be getting what appears to be a DOS, after a day or so when our users try to access OWA externally to our network the page simply does not load, if I check the UAG servers at the time there is nothing going, no high CPU and memory usage and no errors anywhere, event viewer etc.  The only way I can resolve the problem is to restart all the TMG and UAG services and or restart the servers.  It's got to a point now where we've had to schedule a daily reboot of the UAG servers to prevent the problem from occurring.

    We have a call open with Microsoft and it's been with them now for a few months and even they are scratching their heads.  If anyone could give me an idea of where I should start looking that would be great.

    The UAG servers are running on Server 2008 R2 SP1, the build number of UAG is 4.0.1752.10073.

    Any help and or advice would be greatly appreciated.

    Thanks,

    Thursday, November 17, 2011 10:06 AM

All replies

  • We are also experiencing these issues where UAG seems to die for no apparent reason and only a reboot fixes it. We are running UAG in a load balanced environment and it is sometimes difficult to work out which node in the arry has gone down becauise of the lack of anything significant in the event log. We also experience UAG to die when we open web monitor and look at the sessions. This happens on both the completely seperate UAG services we are running, I'm not sure if this might be your issue, I have posted about this recently on this forum. 

    Sorry i haven't got anything else to add but it might be worth checking if there is any correlation between opening web monitor and UAG going down.

    Many thanks,

    Peter

    Thursday, November 17, 2011 11:05 AM
  • Very interesting, we are also running UAG in a load balanced environment, we too cannot determine which node is not accepting connections so we have to reboot all the nodes to clear the error.  What hardware load balancers are you using in front of the UAG's?
    Thursday, November 17, 2011 11:14 AM
  • We are load balancing 2 nodes using UAGs internal load balancing mechanism. It is a bit frustrating that when the UAG service on a particular node dies firstly it is nearly impossible to work out which node has the problem and secondly the internal UAG load balancing doesn't recognise there is a problem and carries on forwarding requests to the 'dead' node! I wonder if there is a way to check whether UAG is running correctly on an individual node?

    Thanks,

    Peter 

    Thursday, November 17, 2011 11:24 AM
  • Just as an update to this:

    A way I've found where I can discover which node has failed is my using TMG and use the filter to log only requests coming from a particular user or ip address that has reported problems with the UAG site. I can then see which node the request is going to and hence where the failure is. Saves rebooting the 'working' node which I've done in the past!

    Peter

    Thursday, March 15, 2012 4:08 PM